Ads by TechWords

See your link here
Receive the latest technology news and information.
Hardware
Computerworld Daily News (First Look and Wrap-Up)
Computerworld Blogs Newsletter
The Weekly Top 10
Cloud Computing
View all newsletters




Privacy Policy
 

Japan aims for world's fastest supercomputer

Goal is machine that would crunch numbers 30 times faster than IBM's Blue Gene

May 31, 2005 12:00 PM ET

TOKYO -- Japan will begin research in June to build a supercomputer capable of crunching numbers about 30 times faster than IBM's Blue Gene/ L, the world's current fastest supercomputer, the Japanese government said today.
Japan's Ministry of Education, Culture, Sports, Science and Technology has established a program with NEC Corp., Hitachi Ltd. and several Japanese universities to develop a supercomputer by as early as 2011 that will be able to perform more than 3 quadrillion floating-point operations per second (FLOPS), or 3 petaFLOPS (PFLOPS), according to Toshihiko Hoshino, information, science and technology director at the ministry's Research and Promotion Bureau.
A computer running at such a speed would race ahead of today's fastest systems, which are typically capable of tens of trillions of calculations per second, or tens of TFLOPS. A TFLOPS is 1 trillion calculations per second, and a PFLOPS is 1,000 TFLOPS.
The computer will be used for biotechnology and nanotechnology research for the benefit of Japanese industry, Hoshino said.
The supercomputer performance record is held today by the $100 million Blue Gene/L at Lawrence Livermore National Laboratory in Livermore, Calif. In March, the computer was clocked at 135.3 TFLOPS by the Linpack benchmark, which puts supercomputer systems through a series of mathematical tests. The computer's theoretical peak performance is 360 TFLOPS, according to IBM.
While the exact performance of Japan's supercomputer has yet to be decided, the government wants it to match or exceed the speeds that it predicts U.S. supercomputers will have at the end of the decade, Hoshino said.
"We predict that Blue Gene/L or its successors will be working at about 3 or 4 PFLOPS around 2010. Our target is to be at least the same speed, or faster," he said.
The development will be split into a three-year research period and a two- to three-year building and assembly period. During the research phase, NEC, which specializes in building supercomputers, will research optical interconnect technologies, while Hitachi will research circuit and transistor designs, Hoshino said.
He declined comment on the budget for the project.
Japan's fastest supercomputer today is NEC's SX6-model Earth Simulator, which clocked a peak performance of nearly 36 TFLOPS in April 2002, according to the Linpack benchmark.

Jump to comments

Hardware

Additional Resources

WHITE PAPER
Approximately 60 percent of data migration projects overrun time or budget, while some fail completely. Download this white paper, "Enhancing Your Chance for Successful Data Migration," to learn the critical steps you need to take to execute a data migration project with minimum cost and risk to your business.
WHITE PAPER
Read the Gartner research note to learn why the TCO of a server-based computing deployment used to deliver all applications to users is around 50% lower than that of an unmanaged desktop deployment.
WHITE PAPER
Economic downturns have a tendency to accelerate emerging technologies, boost the adoption of effective solutions, and punish solutions that are not cost competitive or that are out of synch with industry trends. This IDC White Paper presents the results of an IDC survey of 330 companies in Western Europe, Asia/Pacific and the Americas that measures the receptiveness to Linux and takes into consideration changing views driven by the disruptive economic environment that businesses face today.

White Papers & Webcasts

Faster, Cheaper and Easier to Maintain
Can you afford not to upgrade your servers to today's advanced, energy-efficient technologies?  

Do more with less thru Netcool?
Learn how IBM Tivoli® Netcool® solutions can help service providers streamline their operations, improve responsiveness and reduce costs.  

Effectively Implementing Datacenter Automation
Effectively select and deploy the best datacenter automation solution today!

IDC report: Profitability and OSS Support: A Return on Investment Analysis of IBM Tivoli Netcool
IDC studied 14 mobile and fixed-line service providers that implemented Tivoli® Netcool® and found that IBM Tivoli Netcool can help in big ways.  

Aligning IT to Business: The Rising Importance of Application Delivery Networks
Application Delivery Networking (ADN) will play a vital role in helping enterprises incorporate strategic technologies to achieve business initiatives.

IBM Systems Makeover Analysis for Oracle Environments
This brochure shows how the IBM Systems Makeover Analysis takes a look at your current Oracle hardware infrastructure, then proposes a high-level future...  

Lower your IT costs and risks: Get a server makeover
Find out how a server makeover analysis can help you develop a high-level roadmap for your infrastructure.  

Mitigate Risk, Lower Costs and Improve Network Efficiency
Create a stable IP network that not only meets today's challenges, but is flexible enough to also meet future demands.